The Autism “Treatment” That Wasn’t

How They Keep Solving Autism Before Checking the Data

Every so often, autism gets “solved”.

Not quietly. Not cautiously. Not with a group of knowledgeable researchers sat in a room saying, “this might be something, let’s take our time with it.” It arrives like a film trailer. Big announcement, big confidence, big language. And then, a few months later, a much smaller reality quietly walks in behind it and clears its throat.

The latest example is something called Leucovorin.

Now, if you had not heard of Leucovorin in relation to autism before 2025, that is perfectly normal. It has been around for decades, originally used in things like chemotherapy and certain types of anaemia. It is not new. It is not experimental. It is not some recent scientific breakthrough that suddenly appeared out of nowhere.

What is new is its sudden promotion as something relevant to autism.

It did not gradually earn its place through years of careful, widely agreed research in this space. It was pulled into the autism conversation very quickly, presented at a US government event as a potential treatment, particularly around speech. Not in a cautious, “this might help a small subgroup” sort of way. More in a tone that suggested something meaningful had exploded on the scene.

And if you are a parent, or someone looking for answers, that kind of message lands with significant weight. It sounds like serious progress you might be looking for in your world. It sounds like movement. It sounds like something solid.

But if you have watched this cycle before, it sounds like the beginning of something familiar.

Because by March 2026, the tone had changed. The FDA stepped in and said no, this is not a general autism treatment. The evidence is not strong enough. It is being approved for a very rare condition that can resemble autism, involving a specific issue with how folate is transported in the body.

But something that resembles autism is not the same as autism. If I stand in the shape of a tree, it doesn’t make it so. So is a very different sentence to what people thought they heard.

That is the problem with these things. They tend to start wide and confident, and then narrow themselves once someone properly checks what is actually factual.

And what is genuinely there is important, because Leucovorin is not some purpose-built autism drug. It is an existing medication that someone thought might be useful in a very specific biological scenario. Medicine does that all the time. But there is a very clear difference between “this might help a subgroup” and this is a “treatment”.

That difference usually goes into hiding somewhere between research papers and microphones.

To be fair, this was not pulled from nowhere. There is a scientific idea behind it. Some researchers have looked at whether certain autistic children might have difficulties with how folate reaches the brain. There are things like folate receptor antibodies and rare deficiencies that can interfere with that process. So their thinking becomes, if you support that pathway, maybe you improve certain outcomes, such as around communication.

That is a relatively reasonable hypothesis. There were small studies conducted. One trial looked at 48 children and reported improvements in verbal communication, particularly in a subgroup. But autistic children also turn into autistic adults. Where were they in the study? People who have already passed through the learning process and can offer valid retort?

Regardless, this is where things start to drift.

Because a “signal” in a small study very quickly became “we have found something”. And then, if nobody slows it down, it becomes “we have found the thing”.

The reality is much less exciting I’m afraid.

The evidence is not strong enough to support broad use. That is the central point. The FDA said it. The American Academy of Pediatrics said it. Various clinicians said it, just with slightly more polite wording. One of the larger studies that helped build momentum was later retracted because of issues with the data.

Which is not ideal when you are building a national compliance. Sorry, I meant narrative.

And when you look at the numbers involved, it becomes quite obvious how shaky the foundation is. We are talking about studies involving dozens of children. Not hundreds. Not thousands. Dozens. Yet the level of public confidence suggested something far more definitive.

It is a bit like flipping a coin five times, getting four heads, and announcing you have solved probability. You might be onto something, but you would not want to redesign the system around it just yet.

And this is where it becomes clear that this is not really about Leucovorin at all.

Leucovorin is just the latest example of a pattern that keeps repeating itself around autism. One cause, one explanation, one solution, and usually one person standing slightly outside the system saying they have found what everyone else has missed.

It is neat. It is emotionally satisfying. It gives people something to hold onto in a space that is often confusing and complex.

The problem is, autism is not available to fit into the neat category when it comes to why we exist And I exist proudly. It is not one cause, it is not one pathway, and it is definitely not one solution. So anything that arrives packaged like that should immediately feel a bit suspicious, even if it is neatly wrapped in good intentions. With a high dose of skepticism applied.

We have seen similar things before. Environmental causes suddenly presented as definitive. Everyday products blamed with enormous confidence. Studies cited that, when you actually read them, do not quite say what people think they say. But by the time anyone checks properly, the story has already spread.

Saying “this is complicated, uncertain, and will take years to understand” is not a very compelling message. Saying “we might have found something important” is. One of those gets attention.

People want clarity. They want answers. They want something that reduces uncertainty. So when something appears that looks like an answer, it travels. Even if it is built on small data and a lot of interpretation.

Confidence moves faster than caution. It always has.

Where this leaves us is quite simple. Leucovorin may still have a place in very specific cases in a subgroup that resembles autism (but is not autism). There may be a small, clearly defined group of people where something like this is genuinely helpful. That is worth researching properly. That is how science is supposed to work.

But that is not how it was presented. What was presented sounded much bigger than that. And that is the problem.

Because if you strip all of this back to its simplest form, the whole situation can be summed up in one line.

Leucovorin started as a narrow biological idea, was inflated into a broad public promise, and then collapsed back into what it always should have been.

A small, uncertain, subgroup question.

And that is the part that never seems to change.

Autism does not need another “answer”.

It needs people to stop announcing conclusions before they have earned them. And lean the truth of what autism is. A passive, processing difference that requires acceptance, not eradication.